Comparison of the Earth's shortwave radiation measured by ERB instruments (CERES/GERB) - Z. Peter Szewczyk G. Louis Smith Kory J. Priestley ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Comparison of the Earth’s short-
wave radiation measured by ERB
instruments (CERES/GERB)
Z. Peter Szewczyk
G. Louis Smith
Kory J. Priestley
Pamela E. Mlynczak
ESA/ESTEC Workshop, The Netherlands
Presentation Outline • Special mode of a CERES instrument • Comparing FM1 and FM4 for the ERB dataset • Matching GERB instrument geometry • Comparison of CERES/GERB unfiltered SW radiances for the ERB dataset
Clouds and the CERES
Earth’s Radiant
Energy System
Instrument
• Narrow field-of-view (15x30km at nadir) scanning radiometer:
• Shortwave channel (0.3-5µm),
• Total channel (0.3-100µm),
• Window channel (8-12µm)
• PFM on board TRMM (1998, failed 06/2000)
• FM1 & FM2 on board Terra (in service from 03/2000)
• FM3 & FM4 on board Aqua (in service from 06/2002)
CERES
CERES
Calibration stability monitored with:
• On-board calibration sources (blackbodies, lamps, solar)
• Multi-channel and multi-instrument consistency
• Geophysical validation
Gain drifts can be detected at the 0.1% level, and corrected!
A part of validated data set for the radiation budget since 1984
Terra & Aqua orbits
Aqua
Terra
• Sun-synchronous, inclination angle 98.2o and 81.8o
• Equatorial crossing time: 10:30AM and 1:30PM
• about 15 minutes apart at nodes
CERES normal operation modes • Cross-track (XT) Scan • Rotational Azimuth Plane Scan (RAPS)
Special mode
Programmable Azimuth Plane Scan (PAPS) mode:
• Scanning plane orientation follows a prescribed schedule
Step-wise changes of the azimuth angle
Time and angle changes depend on satellite position in an orbit
• Increases sampling by an order of magnitude
PAPS applications:
• Special observations
Earth targets
Matching viewing geometry of other instruments
Sampling within required scan plane orientationComparing remote sensing
instruments
Comparison of instrument measurements
should be performed
under the most ideal conditionsFM1/FM4 comparison To validate FM4 (Aqua) radiances for ERB dataset Campaign around the summer solstice of ’02 and ‘03 Greenland appears to be the most homogenous: • FM1 (Terra) and FM4 (Aqua) 15 minutes apart • Scans orthogonal to the solar plane • VZA matched within 10o, RAZ within 1o • About 90 sec of data per orbit
FM1 scan over Greenland
Unfiltered shortwave radiances at 13:40 on 07/10/2003
0 100 203
Watts per square meter per steradianFM4 scan over Greenland
Unfiltered shortwave radiances at 13:54 on 07/10/2003
0 100 200
Watts per square meter per steradianFM1 scanning pattern
Unfiltered shortwave radiances on 07/10/2003
0 100 217
Watts per square meter per steradianFM4 scanning pattern
Unfiltered shortwave shortwave on 07/10/2003
0 100 219
Watts per square meter per steradianData processing constraints Direct comparison of radiances: • difference of averages No time issue: • Terra – Aqua 15 minutes apart Spatial noise dominates: • averaging over 1o x 1o grid-boxes • at least 20 footprints or 75% of area covered Matching geometry: • 10o tolerance for the VZA for all three channels • 1o tolerance for the relative azimuth for shortwave
Statistics
Direct comparison of radiances:
• Each orbital crossing is an independent sample
• Difference computed over a grid-box and orbital crossing
• Uncertainty estimated using a 95% confidence level
t# / 2!
α – test: "=
NShortwave radiances
SW radiances over Greenland
Results for Greenland
Mean FM4 Δ mean Δ mean Δσ α−test
Period NorbX
[Wm-2sr -1] [Wm-2 sr -1] [%] [%] [%]
㻭㼘㼏㼜㻏㻃㻕㻓㻓㻕 㻔㻘㻚㻑㻖 㻓㻑㻖㻖 㻓㻑㻕 㻔㻑㻕㻓 㻖㻓 㻓㻑㻗
㻤㼘㼊㼘㼖㼗㻏㻃㻕㻓㻓㻕 㻔㻗㻘㻑㻔 㻓㻑㻕㻔 㻓㻑㻔 㻓㻑㻜㻓 㻔㻖 㻓㻑㻗
㻭㼘㼑㼈㻏㻃㻕㻓㻓㻖 㻔㻘㻙㻑㻜 㻐㻓㻑㻛㻚 㻐㻓㻑㻙 㻓㻑㻙㻓 㻕㻘 㻓㻑㻕FM1/FM4 summary • Comparison procedure for FM1 and FM4 was shown to be well planned and executed • Data analysis fully demonstrated the 1% consistency in radiance measurements (small spread) • Experiment is repeated during the summer solstice every year in an effort to monitor the CERES performance • CERES instruments have delivered a high quality Earth radiation budget (ERB) data set since 1998
GERB measurements
• Geo-stationary Earth Radiation Budget instrument
On board MSG located at 3.4W (10.5 W)
Array of 256 detectors covering the visible portion of the Earth
Short and long-wave radiation image every 15min.CERES/GERB comparison
To validate GERB radiances for ERB dataset
• Campaign around the winter solstice of 2003
PAPS matches GERB viewing geometry
FM2 on Terra in this special mode for about 15 days
4 daytime orbits per day for about 85 min scanning
Unfiltered shortwave radiances are produced by RMIB for this study
GERB pointing accuracy is improved using SEVIRI data
GERB geolocation error is about a half of its footprint without biasCERES/GERB daily scanning
CERES/GERB daily scanning
CERES/GERB daily scanning
CERES/GERB daily scanning
CERES/GERB comparison
Preliminary results for SW
Mean FM2 Δ mean Δ mean
Scene ID Δσ NorbX α−test
[Wm-2sr–1 ] [Wm-2 sr–1] %
Clear ocean 30.8 0.2 0.7 3.2 140 0.37
Clear desert 75.3 0.7 1.0 2.2 42 0.65
Overcast 145.5 - 8.9 - 6.2 7.4 166 1.14
All-sky 81.0 - 2.8 - 4.0 11.3 621 0.92CERES/GERB summary • Preliminary results presented • Large amount of data using the PAPS mode • 1% consistency in the middle of spectrum • More bias for bright scenes • GERB’s improved pointing accuracy is a key to promising results • Further improved geolocation needed!
Concluding remarks • CERES participated in variety of campaigns • PAPS mode for special observations • Planning tools reside on a website • Rapid response due to automation • Free service to the science community http://asd-www.larc.nasa.gov/PAPS/cgi-bin/rygar.cgi
You can also read
